An empirical comparison between stochastic and deterministic centroid initialisation for K-Means variations
This work addresses the initialization sensitivity problem in K-Means clustering for data scientists and practitioners, but it is incremental as it builds on existing variations and techniques.
The study compared deterministic and stochastic centroid initialization methods for K-Means variations, finding that deterministic methods generally perform better, but simpler stochastic methods with multiple runs can yield improved clustering, with execution time considerations making deterministic approaches competitive.
K-Means is one of the most used algorithms for data clustering and the usual clustering method for benchmarking. Despite its wide application it is well-known that it suffers from a series of disadvantages; it is only able to find local minima and the positions of the initial clustering centres (centroids) can greatly affect the clustering solution. Over the years many K-Means variations and initialisation techniques have been proposed with different degrees of complexity. In this study we focus on common K-Means variations along with a range of deterministic and stochastic initialisation techniques. We show that, on average, more sophisticated initialisation techniques alleviate the need for complex clustering methods. Furthermore, deterministic methods perform better than stochastic methods. However, there is a trade-off: less sophisticated stochastic methods, executed multiple times, can result in better clustering. Factoring in execution time, deterministic methods can be competitive and result in a good clustering solution. These conclusions are obtained through extensive benchmarking using a range of synthetic model generators and real-world data sets.